Chapter 1: Test Planning and Results

Watch the Interactive Video: Test Planning Overview

Testing Inferences

Suppose you tested your prototype with independent variable “A” at level a, and an observed dependent variable “B” at level b.

What happens for other levels of A? Did you measure for everything that changed, which may have been more than just A?

Looking back at the studying example, your independent variables of study time and location may affect more than just your test score. For example, they may also affect your homework scores. Could the result simply be due to “random chance”?

Hypotheses

A hypothesis is a testable statement that predicts observable phenomena. It comprises two parts:

• What happens to B when A is changed?
• What happens to B when A is not changed?

A famous hypothesis is that objects of different weights fall at the same rate. A hypothesis we can make about studying is that studying longer with peers will improve your test score.

When evaluating a hypothesis, it can be accepted (true) or rejected (false).

Hypotheses give us testable and systematic explanations for observable phenomena and help to focus our attention during testing. The knowledge gained from testing a hypothesis provides rigorous support for further experiments and design decisions. The data gained can prove or disprove a hypothesis using statistical analysis.

Watch the Video: Test Plan Example

Three Test Plans and Results:  Corn Grinder Prototype Example in ME 270

Click on the link below to download and review three examples of test plans and results using the ME 270 DMADVR toolbox template. The test plan is the first item shown, and scrolling down that same page are the testing results for each of the tabbed tests #1-3. This example is also available Canvas Modules 0 & 00, Module 1 Quiz and Assignments, and Resources.

Three test plans and results examples

Data Analysis

Once the data is collected from the tests, it can be subjected to data analysis. Some common methods of statistical analysis are:

• Measures of central tendency- mean, median, and mode
• Measures of deviation from central tendency- variance, standard deviation
• Regression- Fitting models to data, linear, log-linear, log-log, etc.